R Notebook

Loading the Data

library(readr)

There were 44 warnings (use warnings() to see them)

library(dplyr)

Attaching package: ‘dplyr’

The following object is masked _by_ ‘.GlobalEnv’:

    first

The following objects are masked from ‘package:stats’:

    filter, lag

The following objects are masked from ‘package:base’:

    intersect, setdiff, setequal, union

library(tidyverse)

Registered S3 methods overwritten by 'dbplyr':
  method         from
  print.tbl_lazy     
  print.tbl_sql      
── Attaching packages ───────────────────────────────────────────────────────────────────────────────────────────── tidyverse 1.3.0 ──
✓ ggplot2 3.3.3     ✓ purrr   0.3.4
✓ tibble  3.0.5     ✓ stringr 1.4.0
✓ tidyr   1.1.2     ✓ forcats 0.5.1
── Conflicts ──────────────────────────────────────────────────────────────────────────────────────────────── tidyverse_conflicts() ──
x dplyr::filter() masks stats::filter()
x dplyr::lag()    masks stats::lag()

library(ggrepel)
library(ggplot2)
library(reshape2)

Attaching package: ‘reshape2’

The following object is masked from ‘package:tidyr’:

    smiths

epl_player = read_csv("playerstats_epl.csv")

Duplicated column names deduplicated: 'Progressive Distance' => 'Progressive Distance_1' [83], 'Touches' => 'Touches_1' [100]
── Column specification ──────────────────────────────────────────────────────────────────────────────────────────────────────────────
cols(
  .default = col_double(),
  Date = col_date(format = ""),
  Name = col_character(),
  Round = col_character(),
  Venue = col_character(),
  Result = col_character(),
  Squad = col_character(),
  Opponent = col_character(),
  Start = col_character(),
  Pos = col_character()
)
ℹ Use `spec()` for the full column specifications.

bundesliga_player = read_csv("playerstats_bundesliga.csv")

Duplicated column names deduplicated: 'Progressive Distance' => 'Progressive Distance_1' [83], 'Touches' => 'Touches_1' [100]
── Column specification ──────────────────────────────────────────────────────────────────────────────────────────────────────────────
cols(
  .default = col_double(),
  Date = col_date(format = ""),
  Name = col_character(),
  Round = col_character(),
  Venue = col_character(),
  Result = col_character(),
  Squad = col_character(),
  Opponent = col_character(),
  Start = col_character(),
  Pos = col_character()
)
ℹ Use `spec()` for the full column specifications.

ligue1_player = read_csv("playerstats_ligue1.csv")

Duplicated column names deduplicated: 'Progressive Distance' => 'Progressive Distance_1' [83], 'Touches' => 'Touches_1' [100]
── Column specification ──────────────────────────────────────────────────────────────────────────────────────────────────────────────
cols(
  .default = col_double(),
  Date = col_date(format = ""),
  Name = col_character(),
  Round = col_character(),
  Venue = col_character(),
  Result = col_character(),
  Squad = col_character(),
  Opponent = col_character(),
  Start = col_character(),
  Pos = col_character()
)
ℹ Use `spec()` for the full column specifications.

seriea_player = read_csv("playerstats_seriea.csv")

Duplicated column names deduplicated: 'Progressive Distance' => 'Progressive Distance_1' [83], 'Touches' => 'Touches_1' [100]
── Column specification ──────────────────────────────────────────────────────────────────────────────────────────────────────────────
cols(
  .default = col_double(),
  Date = col_character(),
  Name = col_character(),
  Round = col_character(),
  Venue = col_character(),
  Result = col_character(),
  Squad = col_character(),
  Opponent = col_character(),
  Start = col_character(),
  Pos = col_character()
)
ℹ Use `spec()` for the full column specifications.

laliga_player = read_csv("playerstats_laliga.csv")

Duplicated column names deduplicated: 'Progressive Distance' => 'Progressive Distance_1' [83], 'Touches' => 'Touches_1' [100]
── Column specification ──────────────────────────────────────────────────────────────────────────────────────────────────────────────
cols(
  .default = col_double(),
  Date = col_character(),
  Name = col_character(),
  Round = col_character(),
  Venue = col_character(),
  Result = col_character(),
  Squad = col_character(),
  Opponent = col_character(),
  Start = col_character(),
  Pos = col_character()
)
ℹ Use `spec()` for the full column specifications.

epl_player["Matchweek"] = 0
bundesliga_player["Matchweek"] = 0
ligue1_player["Matchweek"] = 0
seriea_player["Matchweek"] = 0
laliga_player["Matchweek"] = 0
epl_player

Data Cleaning & Expected Goal vs Goal Comparison

Step 1: Aggregate Expected Goal and Goal Values across all Domestic League Games (per 90) Step 2: Remove Players who played less than 900 minutes (or equivalent of 10 games) Step 3: Plot a graph of Expected Goals vs Goals (per 90) with a line intercept = 0 A point that lies on the line represents a player who scores the same number of goals as he is expected to based on the xG model.

epl_player_summary = epl_player %>%
                        group_by(Name) %>%
                        summarise(TotalMinutesPlayed = sum(Min), TotalExpectedGoals = sum(`Expected Goals`), TotalGoals =sum(Goals), PenaltyGoals = sum(`Penalty Kicks Made`), TotalAssists= sum(Assists), TotalExpectedAssists = sum(`Expected Assist`), TotalnPxG = sum(`Non-Penalty Expected Goals`))
epl_player_summary = epl_player_summary %>% filter(TotalMinutesPlayed >= 900)
epl_player_summary["PerformanceGoalScored"] = epl_player_summary["TotalGoals"] - epl_player_summary["TotalExpectedGoals"]
epl_player_summary["GoalsPer90"] = epl_player_summary["TotalGoals"]*(90/epl_player_summary["TotalMinutesPlayed"])
epl_player_summary["ExpectedGoalsPer90"] = epl_player_summary["TotalExpectedGoals"]*(90/epl_player_summary["TotalMinutesPlayed"])
epl_player_summary["NonPenaltyGoals"] = epl_player_summary["TotalGoals"] - epl_player_summary["PenaltyGoals"]
epl_player_summary["NonPenaltyGoalsPer90"] = epl_player_summary["NonPenaltyGoals"]*(90/epl_player_summary["TotalMinutesPlayed"])
epl_player_summary["nPxGPer90"] = epl_player_summary["TotalnPxG"]*(90/epl_player_summary["TotalMinutesPlayed"])
epl_player_summary["AssistsPer90"] = epl_player_summary["TotalAssists"]*(90/epl_player_summary["TotalMinutesPlayed"])
epl_player_summary["xAssistsPer90"] = epl_player_summary["TotalExpectedAssists"]*(90/epl_player_summary["TotalMinutesPlayed"])
epl_player_summary

NA

Plotting Expected Goals vs Actual Goal (per 90) including Penalties

ggplot(data = epl_player_summary, aes(ExpectedGoalsPer90, GoalsPer90, label = Name)) + geom_text_repel(aes(x = ExpectedGoalsPer90, y = GoalsPer90, label = ifelse(ExpectedGoalsPer90 > 0.5 & ExpectedGoalsPer90 > GoalsPer90, as.character(Name),'')), max.overlaps = 10, point.padding = 0.1, nudge_x = 0.1, nudge_y = -0.1, force  = 2, force_pull = 0) + geom_point(aes(color = cut(ExpectedGoalsPer90, c(-Inf, 0.5, Inf)))) +geom_abline(intercept = 0, slope = 1, color = "orange") +xlim(0,1.1) +ylim(0,1.2) + geom_text_repel(aes(x = ExpectedGoalsPer90, y = GoalsPer90, label = ifelse(ExpectedGoalsPer90 > 0.5 & ExpectedGoalsPer90 < GoalsPer90, as.character(Name),'')), max.overlaps = 10, box.padding = 0.1, nudge_x = 0.1, nudge_y = 0.1) + labs(col="xG per 90") + xlab("xG per 90 Minutes") + ylab("Goals per 90 Minutes") + ggtitle("Plot of Outfield EPL Players Goals per 90 min vs xG per 90 min")

Plotting Expected Goals vs Actual Goal (per 90) excluding Penalties

ggplot(data = epl_player_summary, aes(nPxGPer90, NonPenaltyGoalsPer90, label = Name)) + geom_text_repel(aes(x = nPxGPer90, y = NonPenaltyGoalsPer90, label = ifelse(nPxGPer90 > 0.45 & nPxGPer90 > NonPenaltyGoalsPer90, as.character(Name),'')), max.overlaps = 10, point.padding = 0.1, nudge_x = 0.1, nudge_y = -0.1, force  = 2, force_pull = 0, size = 3) + geom_point(aes(color = cut(nPxGPer90, c(-Inf, 0.45, Inf)))) +geom_abline(intercept = 0, slope = 1, color = "orange") +xlim(0,1.1) +ylim(0,1.2) + geom_text_repel(aes(x = nPxGPer90, y = NonPenaltyGoalsPer90, label = ifelse(nPxGPer90 > 0.45 & nPxGPer90 < NonPenaltyGoalsPer90, as.character(Name),'')), max.overlaps = 10, box.padding = 0.1, nudge_x = 0.1, nudge_y = 0.1, size = 3) + labs(col="npxG per 90") + xlab("Non-Penalty xG per 90 Minutes") + ylab("Goals per 90 Minutes") + ggtitle("Plot of Outfield EPL Players Goals per 90 min vs nPxG per 90 min")

Plotting Expected Assists vs Actual Assist

ggplot(data = epl_player_summary, aes(xAssistsPer90, AssistsPer90, label = Name)) + geom_text_repel(aes(x = xAssistsPer90, y = AssistsPer90, label = ifelse(xAssistsPer90 > 0.25 & xAssistsPer90 > AssistsPer90, as.character(Name),'')), max.overlaps = 10, point.padding = 0.1, nudge_x = 0.2, nudge_y = -0.1, force  = 5, force_pull = 0, size=3, label.padding = 0.5, box.padding = 0.5) + geom_point(aes(color = cut(xAssistsPer90, c(-Inf, 0.25, Inf)))) +geom_abline(intercept = 0, slope = 1, color = "orange") +xlim(0,0.75) +ylim(0,0.75) + geom_text_repel(aes(x = xAssistsPer90, y = AssistsPer90, label = ifelse(xAssistsPer90 > 0.25 & xAssistsPer90 < AssistsPer90, as.character(Name),'')), max.overlaps = 10, box.padding = 0.1, nudge_x = 0.2, nudge_y = 0, force  = 5, force_pull = 0, size=3, label.padding = 0.5) + labs(col="xAssists per 90") + xlab("xAssists per 90 Minutes") + ylab("Assists per 90 Minutes") + ggtitle("Plot of Outfield EPL Players Assists per 90 min vs xAssists per 90 min")

Ignoring unknown parameters: label.paddingIgnoring unknown parameters: label.padding

bundesliga_player_summary = bundesliga_player %>%
                        group_by(Name) %>%
                        summarise(TotalMinutesPlayed = sum(Min), TotalExpectedGoals = sum(`Expected Goals`), TotalGoals =sum(Goals), PenaltyGoals = sum(`Penalty Kicks Made`), TotalAssists= sum(Assists), TotalExpectedAssists = sum(`Expected Assist`), TotalnPxG = sum(`Non-Penalty Expected Goals`))
bundesliga_player_summary = bundesliga_player_summary %>% filter(TotalMinutesPlayed >= 900)
bundesliga_player_summary["PerformanceGoalScored"] = bundesliga_player_summary["TotalGoals"] - bundesliga_player_summary["TotalExpectedGoals"]
bundesliga_player_summary["GoalsPer90"] = bundesliga_player_summary["TotalGoals"]*(90/bundesliga_player_summary["TotalMinutesPlayed"])
bundesliga_player_summary["ExpectedGoalsPer90"] = bundesliga_player_summary["TotalExpectedGoals"]*(90/bundesliga_player_summary["TotalMinutesPlayed"])
bundesliga_player_summary["NonPenaltyGoals"] = bundesliga_player_summary["TotalGoals"] - bundesliga_player_summary["PenaltyGoals"]
bundesliga_player_summary["NonPenaltyGoalsPer90"] = bundesliga_player_summary["NonPenaltyGoals"]*(90/bundesliga_player_summary["TotalMinutesPlayed"])
bundesliga_player_summary["nPxGPer90"] = bundesliga_player_summary["TotalnPxG"]*(90/bundesliga_player_summary["TotalMinutesPlayed"])
bundesliga_player_summary["AssistsPer90"] = bundesliga_player_summary["TotalAssists"]*(90/bundesliga_player_summary["TotalMinutesPlayed"])
bundesliga_player_summary["xAssistsPer90"] = bundesliga_player_summary["TotalExpectedAssists"]*(90/bundesliga_player_summary["TotalMinutesPlayed"])
bundesliga_player_summary

NA

Plotting Expected Goals vs Actual Goal (per 90) including Penalties

ggplot(data = bundesliga_player_summary, aes(ExpectedGoalsPer90, GoalsPer90, label = Name)) + geom_text_repel(aes(x = ExpectedGoalsPer90, y = GoalsPer90, label = ifelse(ExpectedGoalsPer90 > 0.5 & ExpectedGoalsPer90 > GoalsPer90, as.character(Name),'')), max.overlaps = 10, point.padding = 0.1, nudge_x = 0.1, nudge_y = 0, force  = 2, force_pull = 0, size = 3) + geom_point(aes(color = cut(ExpectedGoalsPer90, c(-Inf, 0.5, Inf)))) +geom_abline(intercept = 0, slope = 1, color = "orange") +xlim(0,1.6) +ylim(0,1.6) + geom_text_repel(aes(x = ExpectedGoalsPer90, y = GoalsPer90, label = ifelse(ExpectedGoalsPer90 > 0.5 & ExpectedGoalsPer90 < GoalsPer90, as.character(Name),'')), max.overlaps = 10, box.padding = 0.1, nudge_x = 0.1, nudge_y = 0.1, size = 3) + labs(col="xG per 90") + xlab("xG per 90 Minutes") + ylab("Goals per 90 Minutes") + ggtitle("Plot of Outfield Bundesliga Players Goals per 90 min vs xG per 90 min")

Plotting Expected Goals vs Actual Goal (per 90) excluding Penalties

ggplot(data = bundesliga_player_summary, aes(nPxGPer90, NonPenaltyGoalsPer90, label = Name)) + geom_text_repel(aes(x = nPxGPer90, y = NonPenaltyGoalsPer90, label = ifelse(nPxGPer90 > 0.45 & nPxGPer90 > NonPenaltyGoalsPer90, as.character(Name),'')), max.overlaps = 10, point.padding = 0.1, nudge_x = 0.3, nudge_y = -0.2, force  = 3, force_pull = 0, size=3, label.padding = 0.5, box.padding = 0.5) + geom_point(aes(color = cut(nPxGPer90, c(-Inf, 0.45, Inf)))) +geom_abline(intercept = 0, slope = 1, color = "orange") +xlim(0,1.6) +ylim(0,1.6) + geom_text_repel(aes(x = nPxGPer90, y = NonPenaltyGoalsPer90, label = ifelse(nPxGPer90 > 0.45 & nPxGPer90 < NonPenaltyGoalsPer90, as.character(Name),'')), max.overlaps = 10, nudge_x = 0.1, nudge_y = 0.5, size=3, label.padding = 0.5, box.padding = 0.5) + labs(col="npxG per 90") + xlab("Non-Penalty xG per 90 Minutes") + ylab("Goals per 90 Minutes") + ggtitle("Plot of Outfield Bundesliga Players Goals per 90 min vs nPxG per 90 min")

Ignoring unknown parameters: label.paddingIgnoring unknown parameters: label.padding

Plotting Expected Assists vs Actual Assist

ggplot(data = bundesliga_player_summary, aes(xAssistsPer90, AssistsPer90, label = Name)) + geom_text_repel(aes(x = xAssistsPer90, y = AssistsPer90, label = ifelse(xAssistsPer90 > 0.3 & xAssistsPer90 > AssistsPer90, as.character(Name),'')), max.overlaps = 10, point.padding = 0.1, nudge_x = 0.2, nudge_y = -0.2, force  = 5, force_pull = 0, size=3, label.padding = 0.5, box.padding = 0.5) + geom_point(aes(color = cut(xAssistsPer90, c(-Inf, 0.3, Inf)))) +geom_abline(intercept = 0, slope = 1, color = "orange") +xlim(0,0.75) +ylim(0,0.75) + geom_text_repel(aes(x = xAssistsPer90, y = AssistsPer90, label = ifelse(xAssistsPer90 > 0.3 & xAssistsPer90 < AssistsPer90, as.character(Name),'')), max.overlaps = 10, box.padding = 0.5, nudge_x = 0.25, nudge_y = 0, force  = 5, force_pull = 0, size=3, label.padding = 0.5) + labs(col="xAssists per 90") + xlab("xAssists per 90 Minutes") + ylab("Assists per 90 Minutes") + ggtitle("Plot of Outfield Bundesliga Players Assists per 90 min vs xAssists per 90 min")

Ignoring unknown parameters: label.paddingIgnoring unknown parameters: label.padding

ligue1_player_summary = ligue1_player %>%
                        group_by(Name) %>%
                        summarise(TotalMinutesPlayed = sum(Min), TotalExpectedGoals = sum(`Expected Goals`), TotalGoals =sum(Goals), PenaltyGoals = sum(`Penalty Kicks Made`), TotalAssists= sum(Assists), TotalExpectedAssists = sum(`Expected Assist`), TotalnPxG = sum(`Non-Penalty Expected Goals`))
ligue1_player_summary = ligue1_player_summary %>% filter(TotalMinutesPlayed >= 900)
ligue1_player_summary["PerformanceGoalScored"] = ligue1_player_summary["TotalGoals"] - ligue1_player_summary["TotalExpectedGoals"]
ligue1_player_summary["GoalsPer90"] = ligue1_player_summary["TotalGoals"]*(90/ligue1_player_summary["TotalMinutesPlayed"])
ligue1_player_summary["ExpectedGoalsPer90"] = ligue1_player_summary["TotalExpectedGoals"]*(90/ligue1_player_summary["TotalMinutesPlayed"])
ligue1_player_summary["NonPenaltyGoals"] = ligue1_player_summary["TotalGoals"] - ligue1_player_summary["PenaltyGoals"]
ligue1_player_summary["NonPenaltyGoalsPer90"] = ligue1_player_summary["NonPenaltyGoals"]*(90/ligue1_player_summary["TotalMinutesPlayed"])
ligue1_player_summary["nPxGPer90"] = ligue1_player_summary["TotalnPxG"]*(90/ligue1_player_summary["TotalMinutesPlayed"])
ligue1_player_summary["AssistsPer90"] = ligue1_player_summary["TotalAssists"]*(90/ligue1_player_summary["TotalMinutesPlayed"])
ligue1_player_summary["xAssistsPer90"] = ligue1_player_summary["TotalExpectedAssists"]*(90/ligue1_player_summary["TotalMinutesPlayed"])
ligue1_player_summary

NA

Plotting Expected Goals vs Actual Goal (per 90) including Penalties

ggplot(data = ligue1_player_summary, aes(ExpectedGoalsPer90, GoalsPer90, label = Name)) + geom_text_repel(aes(x = ExpectedGoalsPer90, y = GoalsPer90, label = ifelse(ExpectedGoalsPer90 > 0.5 & ExpectedGoalsPer90 > GoalsPer90, as.character(Name),'')), max.overlaps = 10, point.padding = 0.1, nudge_x = 0.1, nudge_y = 0, force  = 2, force_pull = 0, size = 3) + geom_point(aes(color = cut(ExpectedGoalsPer90, c(-Inf, 0.5, Inf)))) +geom_abline(intercept = 0, slope = 1, color = "orange") +xlim(0,1.2) +ylim(0,1.2) + geom_text_repel(aes(x = ExpectedGoalsPer90, y = GoalsPer90, label = ifelse(ExpectedGoalsPer90 > 0.5 & ExpectedGoalsPer90 < GoalsPer90, as.character(Name),'')), max.overlaps = 10, box.padding = 0.1, nudge_x = -0.3, nudge_y = 0, size = 3) + labs(col="xG per 90") + xlab("xG per 90 Minutes") + ylab("Goals per 90 Minutes") + ggtitle("Plot of Outfield Ligue 1 Players Goals per 90 min vs xG per 90 min")

Plotting Expected Goals vs Actual Goal (per 90) excluding Penalties

ggplot(data = ligue1_player_summary, aes(nPxGPer90, NonPenaltyGoalsPer90, label = Name)) + geom_text_repel(aes(x = nPxGPer90, y = NonPenaltyGoalsPer90, label = ifelse(nPxGPer90 > 0.45 & nPxGPer90 > NonPenaltyGoalsPer90, as.character(Name),'')), max.overlaps = 10, point.padding = 0.1, nudge_x = 0.3, nudge_y = -0.2, force  = 3, force_pull = 0, size=3, label.padding = 0.5, box.padding = 0.5) + geom_point(aes(color = cut(nPxGPer90, c(-Inf, 0.45, Inf)))) +geom_abline(intercept = 0, slope = 1, color = "orange") +xlim(0,1.6) +ylim(0,1.6) + geom_text_repel(aes(x = nPxGPer90, y = NonPenaltyGoalsPer90, label = ifelse(nPxGPer90 > 0.45 & nPxGPer90 < NonPenaltyGoalsPer90, as.character(Name),'')), max.overlaps = 10, nudge_x = 0.1, nudge_y = 0.5, size=3, label.padding = 0.5, box.padding = 0.5) + labs(col="npxG per 90") + xlab("Non-Penalty xG per 90 Minutes") + ylab("Goals per 90 Minutes") + ggtitle("Plot of Outfield Ligue 1 Players Goals per 90 min vs nPxG per 90 min")

Ignoring unknown parameters: label.paddingIgnoring unknown parameters: label.padding

Plotting Expected Assists vs Actual Assist

ggplot(data = ligue1_player_summary, aes(xAssistsPer90, AssistsPer90, label = Name)) + geom_text_repel(aes(x = xAssistsPer90, y = AssistsPer90, label = ifelse(xAssistsPer90 > 0.3 & xAssistsPer90 > AssistsPer90, as.character(Name),'')), max.overlaps = 10, point.padding = 0.1, nudge_x = 0.2, nudge_y = -0.2, force  = 5, force_pull = 0, size=3, label.padding = 0.5, box.padding = 0.5) + geom_point(aes(color = cut(xAssistsPer90, c(-Inf, 0.3, Inf)))) +geom_abline(intercept = 0, slope = 1, color = "orange") +xlim(0,1) +ylim(0,1) + geom_text_repel(aes(x = xAssistsPer90, y = AssistsPer90, label = ifelse(xAssistsPer90 > 0.3 & xAssistsPer90 < AssistsPer90, as.character(Name),'')), max.overlaps = 10, box.padding = 0.5, nudge_x = 0.25, nudge_y = 0, force  = 5, force_pull = 0, size=3, label.padding = 0.5) + labs(col="xAssists per 90") + xlab("xAssists per 90 Minutes") + ylab("Assists per 90 Minutes") + ggtitle("Plot of Outfield Ligue 1 Players Assists per 90 min vs xAssists per 90 min")

Ignoring unknown parameters: label.paddingIgnoring unknown parameters: label.padding

seriea_player_summary = seriea_player %>%
                        group_by(Name) %>%
                        summarise(TotalMinutesPlayed = sum(Min), TotalExpectedGoals = sum(`Expected Goals`), TotalGoals =sum(Goals), PenaltyGoals = sum(`Penalty Kicks Made`), TotalAssists= sum(Assists), TotalExpectedAssists = sum(`Expected Assist`), TotalnPxG = sum(`Non-Penalty Expected Goals`))
seriea_player_summary = seriea_player_summary %>% filter(TotalMinutesPlayed >= 900)
seriea_player_summary["PerformanceGoalScored"] = seriea_player_summary["TotalGoals"] - seriea_player_summary["TotalExpectedGoals"]
seriea_player_summary["GoalsPer90"] = seriea_player_summary["TotalGoals"]*(90/seriea_player_summary["TotalMinutesPlayed"])
seriea_player_summary["ExpectedGoalsPer90"] = seriea_player_summary["TotalExpectedGoals"]*(90/seriea_player_summary["TotalMinutesPlayed"])
seriea_player_summary["NonPenaltyGoals"] = seriea_player_summary["TotalGoals"] - seriea_player_summary["PenaltyGoals"]
seriea_player_summary["NonPenaltyGoalsPer90"] = seriea_player_summary["NonPenaltyGoals"]*(90/seriea_player_summary["TotalMinutesPlayed"])
seriea_player_summary["nPxGPer90"] = seriea_player_summary["TotalnPxG"]*(90/seriea_player_summary["TotalMinutesPlayed"])
seriea_player_summary["AssistsPer90"] = seriea_player_summary["TotalAssists"]*(90/seriea_player_summary["TotalMinutesPlayed"])
seriea_player_summary["xAssistsPer90"] = seriea_player_summary["TotalExpectedAssists"]*(90/seriea_player_summary["TotalMinutesPlayed"])
seriea_player_summary

NA

Plotting Expected Goals vs Actual Goal (per 90) including Penalties

ggplot(data = seriea_player_summary, aes(ExpectedGoalsPer90, GoalsPer90, label = Name)) + geom_text_repel(aes(x = ExpectedGoalsPer90, y = GoalsPer90, label = ifelse(ExpectedGoalsPer90 > 0.6 & ExpectedGoalsPer90 > GoalsPer90, as.character(Name),'')), max.overlaps = 10, point.padding = 0.1, nudge_x = 0.1, nudge_y = 0, force  = 2, force_pull = 0, size = 3) + geom_point(aes(color = cut(ExpectedGoalsPer90, c(-Inf, 0.6, Inf)))) +geom_abline(intercept = 0, slope = 1, color = "orange") +xlim(0,1.5) +ylim(0,1.5) + geom_text_repel(aes(x = ExpectedGoalsPer90, y = GoalsPer90, label = ifelse(ExpectedGoalsPer90 > 0.6 & ExpectedGoalsPer90 < GoalsPer90, as.character(Name),'')), max.overlaps = 10, box.padding = 0.1, nudge_x = -0.3, nudge_y = 0, size = 3) + labs(col="xG per 90") + xlab("xG per 90 Minutes") + ylab("Goals per 90 Minutes") + ggtitle("Plot of Outfield Serie A Players Goals per 90 min vs xG per 90 min")

Plotting Expected Goals vs Actual Goal (per 90) excluding Penalties

ggplot(data = seriea_player_summary, aes(nPxGPer90, NonPenaltyGoalsPer90, label = Name)) + geom_text_repel(aes(x = nPxGPer90, y = NonPenaltyGoalsPer90, label = ifelse(nPxGPer90 > 0.5 & nPxGPer90 > NonPenaltyGoalsPer90, as.character(Name),'')), max.overlaps = 10, point.padding = 0.1, nudge_x = 0.3, nudge_y = -0.2, force  = 3, force_pull = 0, size=3, label.padding = 0.5, box.padding = 0.5) + geom_point(aes(color = cut(nPxGPer90, c(-Inf, 0.5, Inf)))) +geom_abline(intercept = 0, slope = 1, color = "orange") +xlim(0,1.5) +ylim(0,1.5) + geom_text_repel(aes(x = nPxGPer90, y = NonPenaltyGoalsPer90, label = ifelse(nPxGPer90 > 0.5 & nPxGPer90 < NonPenaltyGoalsPer90, as.character(Name),'')), max.overlaps = 10, nudge_x = 0.1, nudge_y = 0.5, size=3, label.padding = 0.5, box.padding = 0.5) + labs(col="npxG per 90") + xlab("Non-Penalty xG per 90 Minutes") + ylab("Goals per 90 Minutes") + ggtitle("Plot of Outfield Serie A Players Goals per 90 min vs nPxG per 90 min")

Ignoring unknown parameters: label.paddingIgnoring unknown parameters: label.padding

Plotting Expected Assists vs Actual Assist

ggplot(data = seriea_player_summary, aes(xAssistsPer90, AssistsPer90, label = Name)) + geom_text_repel(aes(x = xAssistsPer90, y = AssistsPer90, label = ifelse(xAssistsPer90 > 0.3 & xAssistsPer90 > AssistsPer90, as.character(Name),'')), max.overlaps = 10, point.padding = 0.1, nudge_x = 0.2, nudge_y =0, force  = 5, force_pull = 0, size=3, label.padding = 0.5, box.padding = 0.5) + geom_point(aes(color = cut(xAssistsPer90, c(-Inf, 0.3, Inf)))) +geom_abline(intercept = 0, slope = 1, color = "orange") +xlim(0,1) +ylim(0,1) + geom_text_repel(aes(x = xAssistsPer90, y = AssistsPer90, label = ifelse(xAssistsPer90 > 0.3 & xAssistsPer90 < AssistsPer90, as.character(Name),'')), max.overlaps = 10, box.padding = 0.5, nudge_x = 0.25, nudge_y = 0, force  = 5, force_pull = 0, size=3, label.padding = 0.5) + labs(col="xAssists per 90") + xlab("xAssists per 90 Minutes") + ylab("Assists per 90 Minutes") + ggtitle("Plot of Outfield Serie A Players Assists per 90 min vs xAssists per 90 min")

Ignoring unknown parameters: label.paddingIgnoring unknown parameters: label.padding

laliga_player_summary = laliga_player %>%
                        group_by(Name) %>%
                        summarise(TotalMinutesPlayed = sum(Min), TotalExpectedGoals = sum(`Expected Goals`), TotalGoals =sum(Goals), PenaltyGoals = sum(`Penalty Kicks Made`), TotalAssists= sum(Assists), TotalExpectedAssists = sum(`Expected Assist`), TotalnPxG = sum(`Non-Penalty Expected Goals`))
laliga_player_summary = laliga_player_summary %>% filter(TotalMinutesPlayed >= 900)
laliga_player_summary["PerformanceGoalScored"] = laliga_player_summary["TotalGoals"] - laliga_player_summary["TotalExpectedGoals"]
laliga_player_summary["GoalsPer90"] = laliga_player_summary["TotalGoals"]*(90/laliga_player_summary["TotalMinutesPlayed"])
laliga_player_summary["ExpectedGoalsPer90"] = laliga_player_summary["TotalExpectedGoals"]*(90/laliga_player_summary["TotalMinutesPlayed"])
laliga_player_summary["NonPenaltyGoals"] = laliga_player_summary["TotalGoals"] - laliga_player_summary["PenaltyGoals"]
laliga_player_summary["NonPenaltyGoalsPer90"] = laliga_player_summary["NonPenaltyGoals"]*(90/laliga_player_summary["TotalMinutesPlayed"])
laliga_player_summary["nPxGPer90"] = laliga_player_summary["TotalnPxG"]*(90/laliga_player_summary["TotalMinutesPlayed"])
laliga_player_summary["AssistsPer90"] = laliga_player_summary["TotalAssists"]*(90/laliga_player_summary["TotalMinutesPlayed"])
laliga_player_summary["xAssistsPer90"] = laliga_player_summary["TotalExpectedAssists"]*(90/laliga_player_summary["TotalMinutesPlayed"])
laliga_player_summary

NA

Plotting Expected Goals vs Actual Goal (per 90) including Penalties

ggplot(data = laliga_player_summary, aes(ExpectedGoalsPer90, GoalsPer90, label = Name)) + geom_text_repel(aes(x = ExpectedGoalsPer90, y = GoalsPer90, label = ifelse(ExpectedGoalsPer90 > 0.5 & ExpectedGoalsPer90 > GoalsPer90, as.character(Name),'')), max.overlaps = 10, point.padding = 0.1, nudge_x = 0.1, nudge_y = 0, force  = 2, force_pull = 0, size = 3) + geom_point(aes(color = cut(ExpectedGoalsPer90, c(-Inf, 0.5, Inf)))) +geom_abline(intercept = 0, slope = 1, color = "orange") +xlim(0,1.0) +ylim(0,1.0) + geom_text_repel(aes(x = ExpectedGoalsPer90, y = GoalsPer90, label = ifelse(ExpectedGoalsPer90 > 0.5 & ExpectedGoalsPer90 < GoalsPer90, as.character(Name),'')), max.overlaps = 10, box.padding = 0.1, nudge_x = -0.3, nudge_y = 0, size = 3) + labs(col="xG per 90") + xlab("xG per 90 Minutes") + ylab("Goals per 90 Minutes") + ggtitle("Plot of Outfield La Liga Players Goals per 90 min vs xG per 90 min")

Plotting Expected Goals vs Actual Goal (per 90) excluding Penalties

ggplot(data = laliga_player_summary, aes(nPxGPer90, NonPenaltyGoalsPer90, label = Name)) + geom_text_repel(aes(x = nPxGPer90, y = NonPenaltyGoalsPer90, label = ifelse(nPxGPer90 > 0.4 & nPxGPer90 > NonPenaltyGoalsPer90, as.character(Name),'')), max.overlaps = 10, point.padding = 0.1, nudge_x = 0.3, nudge_y =0, force  = 3, force_pull = 0, size=3, label.padding = 0.5, box.padding = 0.5) + geom_point(aes(color = cut(nPxGPer90, c(-Inf, 0.4, Inf)))) +geom_abline(intercept = 0, slope = 1, color = "orange") +xlim(0,1) +ylim(0,1) + geom_text_repel(aes(x = nPxGPer90, y = NonPenaltyGoalsPer90, label = ifelse(nPxGPer90 > 0.4 & nPxGPer90 < NonPenaltyGoalsPer90, as.character(Name),'')), max.overlaps = 10, nudge_x = 0.1, nudge_y = 0, size=3, label.padding = 0.5, box.padding = 0.5) + labs(col="npxG per 90") + xlab("Non-Penalty xG per 90 Minutes") + ylab("Goals per 90 Minutes") + ggtitle("Plot of Outfield La Liga Players Goals per 90 min vs nPxG per 90 min")

Ignoring unknown parameters: label.paddingIgnoring unknown parameters: label.padding

Plotting Expected Assists vs Actual Assist

ggplot(data = laliga_player_summary, aes(xAssistsPer90, AssistsPer90, label = Name)) + geom_text_repel(aes(x = xAssistsPer90, y = AssistsPer90, label = ifelse(xAssistsPer90 > 0.25 & xAssistsPer90 > AssistsPer90, as.character(Name),'')), max.overlaps = 10, point.padding = 0.1, nudge_x = 0.2, nudge_y =0, force  = 5, force_pull = 0, size=3, label.padding = 0.5, box.padding = 0.5) + geom_point(aes(color = cut(xAssistsPer90, c(-Inf, 0.25, Inf)))) +geom_abline(intercept = 0, slope = 1, color = "orange") +xlim(0,1) +ylim(0,1) + geom_text_repel(aes(x = xAssistsPer90, y = AssistsPer90, label = ifelse(xAssistsPer90 > 0.25 & xAssistsPer90 < AssistsPer90, as.character(Name),'')), max.overlaps = 10, box.padding = 0.5, nudge_x = 0.25, nudge_y = 0.1, force  = 5, force_pull = 0, size=3, label.padding = 0.5) + labs(col="xAssists per 90") + xlab("xAssists per 90 Minutes") + ylab("Assists per 90 Minutes") + ggtitle("Plot of Outfield La Liga Players Assists per 90 min vs xAssists per 90 min")

Ignoring unknown parameters: label.paddingIgnoring unknown parameters: label.padding

get_pass_chart=function(name) {
  player = epl_player[which(epl_player["Name"] == name),]%>%
  select(Matchweek, Round,Opponent, Min, `Total Passes Completed`, `Total Passes Attempted`, `Short Passes Completed`, `Short Passes Attempted`,`Medium Passes Completed`, `Medium Passes Attempted`,`Long Passes Completed`,`Long Passes Attempted`)
  for (i in 1:38) {
    currmw = paste("Matchweek", toString(i), sep=" ")
    if (!any(player==currmw) ) {
      new_row = data.frame(i, currmw,"NA", 0, 0, 0, 0, 0, 0, 0, 0, 0)
      names(new_row) = c("Matchweek","Round", "Opponent", "Min", "Total Passes Completed", "Total Passes Attempted", "Short Passes Completed", "Short Passes Attempted", "Medium Passes Completed", "Medium Passes Attempted", "Long Passes Completed", "Long Passes Attempted")
      player = rbind(player, new_row)
    } else {
      player$Matchweek[player$Round == currmw] = i
    }
  }
  player = player[order(player$Matchweek),]
  
  temp_player <- melt(player[,c('Matchweek','Short Passes Attempted','Medium Passes Attempted','Long Passes Attempted')],id.vars = 1)
  
  title_value = paste(name, "Pass Chart", sep=" ")
  ggplot(temp_player,aes(x = Matchweek,y = value)) + 
    geom_bar(aes(fill = variable),stat = "identity",position = "dodge") +
    ggtitle(title_value) + 
    labs(col="Types of Passes")
  
}

get_pass_chart("Thiago Alcántara")

get_pressure_chart=function(name) {
  player = epl_player[which(epl_player["Name"] == name),]%>%
  select(Matchweek, Round,Opponent, Min, `Number of times applying pressures in Defensive third`, `Number of times applying pressures in Midfield third`, `Number of times applying pressures in Attacking third`)
  for (i in 1:38) {
    currmw = paste("Matchweek", toString(i), sep=" ")
    if (!any(player==currmw) ) {
      new_row = data.frame(i, currmw,"NA", 0, 0, 0, 0)
      names(new_row) = c("Matchweek","Round", "Opponent", "Min", "Number of times applying pressures in Defensive third", "Number of times applying pressures in Midfield third", "Number of times applying pressures in Attacking third")
      player = rbind(player, new_row)
    } else {
      player$Matchweek[player$Round == currmw] = i
    }
  }
  player = player[order(player$Matchweek),]
  colnames(player)[5] = "Defensive3rd"
  colnames(player)[6] = "Midfield3rd"
  colnames(player)[7] = "Attacking3rd"
  temp_player <- melt(player[,c('Matchweek','Defensive3rd','Midfield3rd','Attacking3rd')],id.vars = 1, value.name = "Pressures", variable.name = "Types of Pressure")
  title_value = paste(name, "Pressure Applied Chart", sep=" ")
  ggplot(temp_player,aes(x = Matchweek,y = Pressures)) +
    geom_bar(aes(fill = `Types of Pressure`),stat = "identity",position = "dodge") +
    ggtitle(title_value) +
    ylim(0, 30)
  
}

# get_pressure_chart("Thiago Alcántara")
get_pressure_chart("Trent Alexander-Arnold")

get_pressure_chart("Andrew Robertson")

get_pressure_chart("Mohamed Salah")

get_pressure_chart("Sadio Mané")

# get_pressure_chart("Roberto Firmino")
# get_pressure_chart("Fabinho")
# get_pressure_chart("Georginio Wijnaldum")
# get_pressure_chart("Bruno Fernandes")
# get_pressure_chart("Marcus Rashford")
# get_pressure_chart("Paul Pogba")
# get_pressure_chart("Scott McTominay")
# get_pressure_chart("Fred")
# get_pressure_chart("Edinson Cavani")
# get_pressure_chart("İlkay Gündoğan")
# get_pressure_chart("Kevin De Bruyne")
# get_pressure_chart("Riyad Mahrez")
# get_pressure_chart("Phil Foden")
# get_pressure_chart("Raheem Sterling")
# get_pressure_chart("Rodri")